TLDR: A study engaged diverse communities to identify four key ethical and socio-technical questions for real-world wellbeing robot deployment: safety, who they’re built for, data ownership, and the fundamental purpose of using a robot. The findings highlight the need for user-centered design, transparency, and robust policies to ensure these technologies are developed responsibly and align with public interest.
As robotic technology advances, the idea of robots assisting with our mental wellbeing is becoming a reality. These ‘wellbeing robots’ are designed to support individuals through practices like positive psychology and mindfulness. However, moving these robots from research labs into our daily lives raises many important ethical and practical questions. A recent study, titled “WHO OWNS THE ROBOT ?: F OUR ETHICAL AND SOCIO -TECHNICAL QUESTIONS ABOUT WELLBEING ROBOTS IN THE REAL WORLD THROUGH COMMUNITY ENGAGEMENT”, delves into these concerns by engaging directly with various communities.
Led by researchers Minja Axelsson, Jiaee Cheong, Rune Nyrup, and Hatice Gunes, the study adopted a community-centered approach. Instead of just experts, they sought insights from people often underrepresented in robotics development. This included members of the public at a science festival, women computer scientists, and humanities researchers interested in the history and philosophy of science. Through workshops and interviews, participants reflected on the ethical issues of robotic wellbeing coaches, using a tool called the Social Robot Co-Design Canvas on Ethics.
The researchers analyzed the discussions and identified four overarching ethical and socio-technical questions that roboticists should consider during the development and deployment of wellbeing robots. These questions serve as a framework to ensure that robot development aligns with public interest and addresses potential societal impacts.
Is the robot safe and how can we know that?
One of the primary concerns raised by participants was safety, encompassing both physical and, more significantly, psychological and emotional aspects. People worried about robots potentially causing physical harm, especially to vulnerable groups like children, or even being damaged by upset users. More profoundly, there were concerns about forming inappropriate emotional attachments to robots, leading to disappointment when affection isn’t reciprocated, or over-disclosing sensitive personal information to a non-human entity. Participants emphasized the need for rigorous testing across diverse populations, clear safety requirements, and robust standards to ensure these robots are truly safe for use.
Who is the robot built for and with?
This question addresses issues of equality, cultural sensitivity, accessibility, and inclusivity in robot design. Participants highlighted the problem of AI bias, where robots trained on unrepresentative data might provide inappropriate advice or perpetuate stereotypes. They pointed out the common “whiteness” and gendered voices of robots, questioning whose norms are being reflected in their design. The need for diverse data and design teams was a strong recommendation. Accessibility was another key theme, covering physical accommodations for people with disabilities and the economic affordability of these technologies. The study also noted a potential disconnect between engineering-focused developers, who prioritize functionality, and users who value aesthetics and emotional connection, underscoring the need for human-centered and inclusive design processes.
Who owns the robot and the data?
Privacy and data ownership emerged as critical concerns, particularly given the sensitive nature of wellbeing data. Participants questioned who would have access to their personal stories and health information, and how this data would be stored and used. They raised doubts about whether existing data protection policies, like GDPR, adequately cover social AI systems that interact in human-like ways. The concept of “robot ownership” was often linked to data ownership, with participants expressing discomfort if an employer, for example, owned a wellbeing robot and had access to their data, fearing surveillance or forced wellness. Transparency about data collection, storage, and usage was deemed essential for building trust.
Also Read:
- Designing Empathetic Robots for Social Care: A New Approach to AI in Complex Settings
- The Physical AI Revolution: Policy Gaps and Urgent Actions for Embodied AI
Why a robot?
Despite acknowledging potential benefits, communities expressed a fundamental uncertainty about the necessity of using robots for wellbeing. The discussion revolved around the use of human-like robots in intimate settings, with some finding anthropomorphic features appealing, while others worried about manipulation, deception, and the perpetuation of human biases. Participants questioned the appropriate role of robots – as aids to therapy rather than replacements for human therapists – and warned against over-reliance. They stressed the importance of transparency regarding a robot’s capabilities and limitations, suggesting features like FAQs where robots could directly explain what they are, how they work, and what they cannot do, to manage user expectations and mitigate ethical risks.
In conclusion, this research provides a vital framework for roboticists, urging them to engage in critical reflection and dialogue with communities throughout the design and deployment process. By proactively addressing these four questions – safety, inclusivity, data ownership, and the fundamental purpose of the robot – developers can work towards creating wellbeing robots that are not only technologically advanced but also ethically sound and truly beneficial to society.
